Finite Element Analysis and Failure Mode Characterization of Pyramidal Fin Arrays Produced by Masked Cold Gas Dynamic Spray

Abstract

This work evaluates the shear strength of pyramidal fin arrays made from various feedstock materials (cylindrical aluminum, spherical nickel, and cylindrical stainless steel 304 powders) deposited on an Al6061-T6 substrate. Higher shear strength was measured for the nickel fin array followed by the stainless steel 304 and the aluminum arrays. Different failure modes were observed by inspecting the fracture surfaces under Scanning Electron Microscope. Deposition between the cold sprayed nickel and stainless fins was detected whereas dimples were noticed on the substrate between the fins when aluminum is used as the feedstock material. A numerical simulation of normal and angled impacts using the high strain rate Preston-Tonks-Wallace model was carried out in order to have a better understanding of the experimental results. The equivalent plastic strain (PEEQ) obtained from the finite element analysis at normal impact correlates with the different shear strengths measured experimentally. Furthermore, even if a higher PEEQ was observed for angled impacts compared to its normal collision counterpart, it is suggested that the particles may not bond because of the rotational restitution momentum caused by the tangential friction generated during angled impacts. This rotational restitution momentum was not detected for particle impacts normal to the substrate surface.

Asymmetrical bonding in cold spraying of dissimilar materials

“…The particle velocity measured using the CSM for N 2 -Sprayed Ti was 456±70. The particle velocity measured for Ni powder was 459±61 m/s [36]. For that reason, the particle initial velocity was set to 500 m/s in the model, for both Ni/Ti and N 2 -sprayed Ti/Ni.…”

“…The hydrodynamic stress of elastic response of the impacting particles and substrates were modelled using the Mie-Grüneisen equation of state (EOS) in the linear Us-Up Hugoniot form [36]:…”

“…where p is the hydrodynamic stress, The deviatoric stress of elastic response was assumed to be linear [36]:…”

“…Finite element modelling (FEM) has been extensively used for enhanced understanding of deformation and bonding mechanisms in CS. Different dynamic hardening models have been used to compute the flow stress of high strain rate deformations such as Johnson-Cook (JC) [8,23,25,[31][32][33][34], Preston-Tonks-Wallace (PTW) [15,35,36], Khan-Huang-Liang (KHL) [35] and Zerilli and Armstrong (ZA) [35]. A comparative study on the simulation of single copper particle impact on copper substrates found that the PTW model predicted the deformation fairly accurately [35].…”

“…A comparative study on the simulation of single copper particle impact on copper substrates found that the PTW model predicted the deformation fairly accurately [35]. In addition, the PTW model has been successfully used for simulation of single impact deformation of materials such as aluminium [15], stainless steel (SS) 304 and nickel [36] in CS process. Bonding has been attributed to the thermal softening of the local area and intensified plastic deformation occurring at impact velocities around or beyond the critical velocity of the material [25,26].…”